Stories by tedsanders

Why SWE-bench Verified no longer measures frontier coding capabilities

METR estimates that GPT-5.2 has a 50%-time-horizon of around 6.6 hrs

GPT-5.1 for Developers

GPT-5.1: A smarter, more conversational ChatGPT

OpenAI reasoning system scores 12/12 at the 2025 ICPC World Finals

ChatGPT Sent Me to the ER

Stargate Project: SoftBank, OpenAI, Oracle, MGX to build data centers

Evaluating frontier AI R&D capabilities of LLM agents against human experts

The Zipper Merge (2017)

What Makes Documentation Good

Apple is putting ChatGPT in Siri for free later this year

What Does the Public in Six Countries Think of Generative AI in News? [pdf]

What Makes Documentation Good

Roots of Disagreement on AI Risk

How many legs do ten elephants have, if two of them are legless?

Is the reversal curse in LLMs real?

I came second out of 999 in the Salem Center prediction market tournament

Robotaxis Will Go Through $100B in Losses to Reach Profitable Scale

Transformative AGI by 2043 is <1% likely

1